« Previous | Contents | Next »
Listen
Natural Flood Storage and Extreme Flood
Events Final Report
10 CONCLUSIONS
10.1 'Natural' flood storage
The concept of 'natural flood storage', as used in the
objectives of this project, is confusing. One of the
natural effects of the floodplain is to generate
attenuation during out-of-bank flows as a result of
resistance acting as a hydraulic control (including surface
roughness and discrete roughness elements like hedges and
walls). There may also be distinct controls such as steps
or constrictions in topography. In some cases, these
controls may cause a volume of water to be held for some
time on the floodplain, a process that could be regarded as
a form of 'storage'.
However, these natural sources of attenuation are by
definition present in any flood event that causes a
T-year flow downstream. It is therefore not
appropriate to discuss the size of an event that can be
'contained by natural storage'. Any natural attenuation
caused by a form of natural flood storage, as discussed
above, will be built into the downstream flows for a given
flood event.
It is more useful to consider how the natural floodplain
might be used to provide additional storage-based
attenuation. This would rely on some form of management to
create impoundments by introducing controls on flow and
adding flood banks. The analysis presented in this report
attempts to quantify the natural flood extent for different
events and to investigate the potential to make greater use
of this area for storage. The analysis is broad scale and
does not seek to represent how this storage could be
created in practice; this would be the scope of a
feasibility study in a specific catchment.
The analysis is an attempt to set up a generalised
method that could be applied to any catchment (assuming
there is a DEM), with or without detailed channel data. It
does not try to identify locations for engineered storage
solutions, but could help to show broadly where the natural
floodplain could be used, both in terms of providing the
required volumes and in terms of economic costs
(agricultural damages). Guidance is also given on other
environmental considerations.
10.2 Further steps
The methods suggested in this report are a starting
point for generalised analysis of storage on the natural
floodplain. There are a number of further options that it
has not been possible to explore, but which would be
logical next steps.
The analysis here has concentrated on the 100 year and
200 year flood events. We have considered the storage
needed to reduce a 100 year event to a 5 year flow by using
floodplain normally flooded during a 200 year event. The
flood extents have been generated using a 2-D flood
inundation model in combination with a 1-D routing model.
Future development should look at alternative and more
meaningful ways of defining the potential natural
floodplain area, perhaps by simulating a 1000-year flood
extent or by incorporating knowledge of historical events
in setting out the floodplain outline. It would also be
useful to adjust outlines to exclude areas that could not
be used for storage, such as buildings and infrastructure.
This report sets out the data sources to be considered.
Rather than using the simplified method of applying a
notional average depth over incremental upstream floodplain
areas, future approaches should seek to identify parcels of
land having good potential. This would require some type of
search procedure. One possibility is to aggregate a flood
depth grid into increasingly large spatial units, and to
identify the units having the largest volume at each
scale.
10.3 Potential for use of 'natural'
storage
The results presented for the three study catchments
where models have been run give some indication about
whether it may be viable to provide enough flood storage to
limit a 100 year event to a 5 year downstream flow solely
by using the natural floodplain, as defined by the 200 year
extent. For the White Cart and the South Esk, the results
suggest that such an option may be possible, with average
depths of less than 2.0m being implied within the first few
kilometres of the stream network above the risk location.
For the Tay catchment, the indications are that this type
of flood storage would be less viable.
It is difficult to generalise from these three case
studies. However, a factor in the Tay is that there is
little difference in areal extent between the modelled 100
and 200 year flood outlines in the lower reaches, owing to
the large volumes of flood waters in this large catchment
and the relatively flat valley floor. This limits the
additional storage available in the lower reaches without
requiring significant increases in average depth. This type
of catchment would therefore seem to offer less potential
for additional 'natural' flood storage, although this is
not to say that schemes in suitable locations upstream
would not provide a useful contribution to flood
management.
The effective utilisation of the floodplain for
mitigation of extreme events may however still, in many
cases, require large scale engineering works (if depths of
the order of a metre are required), which would clearly not
be natural features of the landscape and should be designed
appropriately (see, for example, Figure 10-1 showing
flooded washlands on the Middle Rother in England).
Although there may be potential to create wetlands from
these areas, it has to be kept in mind that a wetland area
already contains water prior to flooding, and hence is not
'available storage volume'.
Figure 10-1: Washlands on
the River Rother
Any flooding on agricultural land produces an economic
impact in terms of lost crop (if the land is in arable or
horticultural production) or reduced grass productivity
(for meat, milk, wool or forage production). The degree of
impact depends on the extent, duration, frequency and
timing of the flood event. The economic analysis undertaken
in this project for agricultural production areas has shown
that any flooding of arable and horticultural crops
significantly increases the total economic cost of the
flood due to the high value of these crops and the direct,
operational and clean-up costs borne by the landowner if
the crop is completely lost. Grass based agricultural
systems do suffer productivity losses during the duration
of the flood and any period of sustaining soil saturation
afterwards, but would not suffer a complete crop loss. If
enhanced floodplain inundation is to be actively pursued,
such as providing the landowners affected with an
appropriate compensation payment to have their land
flooded, then arable and horticultural production in these
areas may become uneconomic if the frequency of flooding is
increased significantly and the potential risk of crop loss
increases. These areas would be converted to less intensive
grassland systems. However, the conversion of arable land
to grassland systems may provide further opportunities for
environmental enhancement through biodiversity and water
quality improvements.
Any proposed scheme to enhance the natural storage must
take due regard for all natural and man-made assets on the
floodplain affected. The generic assessment of natural
floodplain storage produced by this project provides
guidance on how all the appropriate datasets should be
reviewed and stakeholder consultations undertaken to ensure
that all the environmental, economic and social issues are
given adequate consideration.
The managed use of natural floodplains for attenuation
of very big floods with long return periods cannot provide
concomitant benefits for biodiversity. By their nature, big
floods occur infrequently and would not provide the regular
inundation (usually at least yearly) required to promote
changes to existing biodiversity (particularly where land
is currently in productive agricultural use).
The environmental effects of providing the substantial
structure(s) required to hold back the flood waters, which
would be used very infrequently, are difficult to define
without reference to a specific scheme. However, it is
likely that the key benefits would be socio-economic (in
terms of reduction in risk to humans) rather than
ecological, which would probably be at best negligible
except in very localised areas.
« Previous | Contents | Next »